You are on page 1of 43

The Office of Public Services Reform

The Drivers of Satisfaction with Public


Services

Research Study Conducted for


the Office of Public Services Reform

April - May 2004


The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Contents
Introduction 1
Executive Summary 3
Key Drivers of Satisfaction 6
Performance on Key Drivers 11
Additional Key Drivers Analysis 26
Expectations of the Public and Private Sector 28
Methods of Contact 32
Conclusions 33

Appendices
Note on weighting
Statistical reliability
Social class definitions
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Introduction
This report contains the findings of research study conducted by the MORI Social
Research Institute on behalf of the Office of Public Service Reform (OPSR) at the
Cabinet Office.

MORI SRI has conducted a review of approaches to measuring and understanding


customer satisfaction with public services, which summarised existing research and
literature on the subject for the OPSR.

This report builds on the body of research reviewed, and ultimately aims to develop
a list of key drivers of satisfaction across a range of services, which could be used in
general customer satisfaction surveys carried out in different public services.

Objectives

In order to improve the quality of public services, it is important to understand


what elements are most important in determining satisfaction with a public service
transaction – the key drivers. While there have been some studies in individual
departments and sectors to identify these factors, little has been done in this country
to understand the generic drivers of satisfaction across the public sector. The main
objective of this research was to identify a set of elements that might have an
impact on satisfaction with different public services, and to use the experiences of a
representative sample of the general public to ask whether a generic set of key
drivers can be found.

In addition, the survey also asked overall attitudes towards the public sector, and
comparisons with the private sector, and measured expectations of response times,
which can be tracked over time.

Methodology
MORI interviewed a representative sample of 1,502 members of the general public
aged 16+. This includes 287 booster interviews about education. The sample for the
survey was drawn using Random Digit Dialling (RDD) and results were recorded
using Computer Assisted Telephone Interviewing (CATI). Fieldwork for the survey
took place between 23 April – 5 May 2004.

Quotas
Quotas were set on age, gender, country/region and work status. Data was weighted
by ethnicity, social class and working status to reflect the known profile of Great
Britain. Additional weighting was carried out on part of the data to correct for extra
interviews about education. (Please refer to the Note on Weighting section
appended to this report for more details.)

1
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Presentation and Interpretation of the Data


Because a sample, not the entire population of Great Britain has been interviewed,
all results are subject to sampling tolerances. This means that both are accurate to
within certain limits, and that not all differences between sub-groups are statistically
significant. (Please refer to the Statistical Reliability section appended to this report
for more details).

Where percentages do not add up to 100, this may be due to multiple responses,
computer rounding or the exclusion of ‘Don’t know/Not stated’ responses.
Throughout the report an asterisk (*) denotes a value of less than half a per cent,
but more than zero.

Publication of Data
Our standard Terms and Conditions apply to this, as to all studies we carry out.
Compliance with the MRS Code of Conduct and our clearing of any copy or data
for publication, web-siting or press release which contains any data derived from
MORI research is necessary. This is to protect our client’s reputation and integrity
as much as our own. We recognise that it is in no one’s best interests to have
survey findings published which could be misinterpreted, or could appear to be
inaccurately, or misleadingly, presented.

2
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Executive Summary
Key Drivers of Satisfaction
This research has identified a powerful set of five factors that drive satisfaction
across public services. This model should be seen as a tool to help all those
involved in delivering public services identify and monitor the areas that are
important to their customers (the techniques used in this study are also
transferable). Individual public services will also have different elements they will
need to cover, but these five factors should be applicable to all. The key drivers, in
order of their impact, are:

1. Delivery - the service delivers the outcome it promised and manages to deal
with any problems that may arise.

2. Timeliness – the service responds immediately to the initial customer


contact and deals with the issue at the heart of it quickly and without
passing it on between staff.

3. Professionalism – staff are competent and treat customers fairly.

4. Information - the information given out to customers is accurate and


comprehensive and they are kept informed about progress.

5. Staff attitude - staff are friendly, polite and sympathetic to customers’ needs.

Key drivers of satisfaction

Main elements Drivers


The final outcome
The way the service kept its promises Delivery
The way the service handled any problems
30
%
Initial wait
How long it takes overall Timeliness
Number of times had to contact the service 24
%
Accuracy
Comprehensiveness Information 18% Satisfaction
Being kept informed about progress
with service
%
Competent staff 16
Being treated fairly Professionalism
%
12

Polite and friendly staff


How sympathetic staff were to your needs Staff attitude

This model explains 67% of the


variation in satisfaction Source: MORI

Among people who visit offices in person, there is a sixth key driver – the physical
environment (whether offices are clean and comfortable, and the appearance of
staff). However, it is not as strong as the five identified above. Access is also a

3
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

factor affecting people’s satisfaction with public services (made up of the choice of
different ways there is to access a service and how easy it is to access a service,
including opening hours). However, it does not come out as a significant driver of
satisfaction in this analysis.

We are also able to identify the key drivers of satisfaction with health services and
education specifically. In both cases, the same five key drivers, in the same order,
are important in driving customer views.

Performance
Performance on key drivers, reflecting overall satisfaction, tends to be high –
especially on ratings of staff. However, satisfaction with public service delivery and
timeliness, the top two key drivers, is relatively lower than on the other service
aspects.

In particular, areas that are crucial to customers which currently receive lower
satisfaction scores are the service’s ability to keep customers ‘informed about progress’
and the way the service ‘handles problems’. Information provision is traditionally an
area where public services could improve, but what this highlights is the importance
of keeping people up-to-date throughout the process, not just providing them with
information at the initial contact. Additionally, over one in five people are
dissatisfied with the ‘overall time’ it took to deal with the reason of contact.

Satisfaction with public services tends to be highest among:

• users of health and education services;

• older people; and

• those who contact services in person.

This is not unusual; health and education services often receive high ratings, and
older people tend to be more satisfied with services generally. There is also some
overlap here: older people are more likely to contact health services, and less likely
to use the phone.

For different reasons, the most dissatisfied groups are found among:

• the middle classes (higher expectations); and

• those who have contacted public services by letter (more likely to be


complaining).

Local government is the worst performing service by some margin.

4
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Expectations of the public and private sector


Comparing the public and private sector, generally people think the private sector
performs better at providing services (particularly men, younger people and the
middle classes), and have higher expectations in terms of response times. There is
some evidence from this survey that people expect lower response times from public
services than they did five years ago.

Contacting public services


Personal contact and telephone are the most frequent and preferred methods of
contact with public services. More people would prefer to use telephone to contact
public services than is currently the case (though it should be noted that people who
use the phone tend to be more dissatisfied, so increasing the use of telephone
contact may have some implications). Nevertheless, it will remain crucial to offer a
range of methods of contact – for example, people tend to prefer the telephone
when contacting health services, the police and local councils, but would rather
contact education and transport services in person.

Conclusions
As has been discovered in other countries, there is a coherent framework of key
drivers of satisfaction across public services (even if individual services will have
their own particular drivers as well). This framework should provide a useful tool
for public service managers and policy makers to focus their efforts on improving
customer service, and to demonstrate what needs to be monitored to evaluate the
impact of these changes on the customer experience.

©MORI/J22127 Ben Page


Gideon Skinner
Darcy Vasickova

5
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Drivers of Satisfaction


Memorable Public Services Experience
On average, people have contacted six public services in the past 12 months. When
asked to pick a particularly memorable experience (either good or bad), half choose
a health service and they were mostly satisfied with the service they received.
Around one in ten choose a memorable contact with a local government service
(such as local council or local leisure centres) or identifies education services. Other
frequently used services are the police, transport and post offices.

Memorable public services experience


Health services 50%
Education services 10%
Your local council 7%
The police 7%
Transport 7%
Post Office 4%
Inland Revenue 3%
Library 3%
Social Services 2%
Local authority sports/leisure centre 2%
Council housing services 2%
Job Centre Plus 1%
Passport Agency 1%

Other 1%
Base: All (1,502) Source: MORI

Overall satisfaction with public services


Three times as many people are satisfied with their memorable experience of public
services than dissatisfied. Perhaps because we asked people to think of a memorable
experience, feelings tend to be concentrated on the more extreme ends of the
satisfaction scale – more are very satisfied or very dissatisfied than just satisfied or
dissatisfied.

Dissatisfaction tends to be higher among young people, the middle classes and
people living in London.

6
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Overall satisfaction
Q Overall, how satisfied or dissatisfied were you with the service
you received?

Very dissatisfied
14%
Dissatisfied
9%
Neither satisfied nor 3% 48% Very satisfied
dissatisfied

26%
Satisfied

Base: All respondents (1,502) Source: MORI

Those with health or education related experiences are significantly more likely to
be satisfied with the service they received than those who are referring to the police,
local council or transport.

Overall satisfaction with the public service used


Q Overall, how satisfied or dissatisfied were you with the service you
received?
Very dissatisfied Dissatisfied Satisfied Very satisfied % saying the
most memorable
experience

All services 14% 9% 26% 48% 100%

Health services 8% 8% 25% 57% 50%

Education services 6%7% 30% 56% 10%

The Police 29% 7% 28% 32% 7%

Local council 36% 14% 20% 21% 7%

Transport 28% 11% 33% 21% 7%


Base: All services (1,502): Health (612), Education (389), Police (87), Local council (84),
Transport (85) Source: MORI

7
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key drivers of satisfaction


The main aim of the survey was to discover what elements of the public service
experience are most closely associated with overall satisfaction. Based on previous
research in this area, altogether 24 individual service aspects were identified and
tested for satisfaction and importance among the general public.

MORI used statistical techniques to collapse these 24 elements into seven common
factors that apply to experiences with the public sector. We then used further
analysis to identify the five key drivers of customer satisfaction across the public
sector:

1. Delivery - the service delivers the outcome it promised and manages to deal
with any problems that may arise.

2. Timeliness – the service responds immediately to the initial customer contact


and deals with the issue at the heart of it quickly and without passing it on
around staff.

3. Professionalism – staff are competent and treat customers fairly.

4. Information - the information given out to customers is accurate and


comprehensive and they are kept informed about progress.

5. Staff attitude - staff are friendly, polite and sympathetic to customers’ needs.

This is a strong model, as it explains 67 % of the variation in overall satisfaction


with public services.

Key drivers of satisfaction

Main elements Drivers


The final outcome
The way the service kept its promises Delivery
The way the service handled any problems
30
%
Initial wait
How long it takes overall Timeliness
Number of times had to contact the service 24
%
Accuracy
Comprehensiveness Information 18% Satisfaction
Being kept informed about progress
with service
%
Competent staff 16
Being treated fairly Professionalism
%
12

Polite and friendly staff


How sympathetic staff were to your needs Staff attitude

This model explains 67% of the


variation in satisfaction Source: MORI

8
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

For people who visit offices in person, there is a sixth key driver – the physical
environment. However, the five key drivers identified above are still the most
important factors to take into account.

Key drivers of satisfaction – personal contact only


Main elements Drivers
The final outcome
The way the service kept its promises Delivery

28
The way the service handled any problems

%
Initial wait
How long it takes overall Timeliness 21
Number of times had to contact the service %

Accuracy
Comprehensiveness Information 18%
Being kept informed about progress Satisfaction
with service
Polite and friendly staff 14%
How sympathetic staff were to your needs
Staff attitude

%
13
Competent staff
Being treated fairly Professionalism

6%
Clean and comfortable premises
The appearance of staff Physical environment

This model explains 66% of the


variation in satisfaction Source: MORI

9
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Reasons for satisfaction and dissatisfaction


When asked to explain why they are satisfied or dissatisfied with the service they
received, the five key factors identified above compare well with what people
express in their own words. The key four issues on top of people’s minds are the
overall service, its efficiency, quality of staff and the information provided to
customers.

Reasons for service satisfaction


Q Why were you satisfied? Q Why were you dissatisfied?

% reason for satisfaction % reason for dissatisfaction

Service
Excellent/good service 43%
Appalling/poor service 41%

Efficiency/Timeliness
Quick/prompt/efficient service 32%
Long/slow/inefficient service 30%

Staff
Good/polite/helpful/kind/caring/friendly staff 30%
Poor/rude/unhelpful/incompetent/bad attitude 21%
Information
Answered all queries/provided good information 12%
Poor communication/contact
7%

Base: All satisfied with service (1143). All dissatisfied with service (311). Source: MORI

10
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Performance on Key Drivers


Key Driver 1: Delivery
Three quarters are satisfied with the outcome of their contact, but fewer are
satisfied with the way the service kept its promises and handled any problems.
Handling problems in particular is one of the lowest ratings out of all the different
aspects measured. Other research has shown that ‘learning from mistakes’ is a key
driver of trust in public services.

Delivery
Q How satisfied or dissatisfied were you with..?

% Dissatisfied % Satisfied

The final outcome 15% 74%

The way the service kept


its promises 15% 69%

The way the service


handled any problems 16% 65%

Base: All respondents (1,502) Source: MORI

11
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

People who contact the service in person tend to be most satisfied, as are those who
contact health and education services. Younger people and those using telephone or
a letter to contact public services are most dissatisfied.

The final The way the The way the


outcome service kept its service
promises handled any
problems
Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis-
Base fied fied fied
% % % % % %
Type of service
Health services (612) 79 11 77 8 74 10
Education services (389) 82 9 76 10 71 10
Local council (84) 58 23 40 43 45 35
The Police (87) 54 35 52 25 53 30
Transport (85) 54 35 52 31 38 31
Method of contact
Personally (812) 78 13 73 11 70 12
Telephone (525) 69 21 64 22 61 22
Letter (78) 64 11 60 23 48 25
Email (39) 83 11 66 17 55 22

Source: MORI

12
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Driver 2: Timeliness


Most customers are satisfied with the speed the public service deals with them.
However, more effort may need to be put into the overall time it takes to get things
done, as this is the area of highest dissatisfaction among all 24 service aspects
measured.

Timeliness
Q How satisfied or dissatisfied were you with..?

% Dissatisfied % Satisfied

How long you had to wait


before someone responded to 18% 75%
you initially

How long it took overall to


deal with the reason you 21% 71%
contacted

The number of times you had to


get in touch with the service 16% 70%
before somebody could deal
with the reason for your contact

Base: All respondents (1,502) Source: MORI

13
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Older people are more satisfied with these aspects of service delivery, as are those
who deal with health or education services and those who made the contact in
person. Those using a letter to contact public services are most dissatisfied, together
with those using local councils which receive the lowest scores for timeliness.

How long you How long it The number of


had to wait took overall to times you had
before deal with the to get in touch
someone reason you with the service
responded to contacted before
you initially somebody
could deal with
the reason for
your contact
Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis-
Base fied fied fied
% % % % % %
Type of service
Health services (612) 80 15 78 16 78 9
Education services (389) 82 10 81 11 80 9
Local council (84) 56 38 45 47 48 45
The Police (87) 69 24 58 34 61 28
Transport (85) 52 23 53 23 46 17
Method of contact
Personally (812) 78 14 75 15 72 11
Telephone (525) 72 22 67 28 69 23
Letter (78) 64 28 58 31 61 28
Email (39) 66 20 54 27 64 23

Source: MORI

14
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Driver 3: Information


Most people are satisfied with the information they receive, but nearly one in five
are dissatisfied with how the service keeps them informed about progress, one of
two areas of lowest satisfaction among all 24 aspects (the other being the overall
time it took to deal with the reason of contact).

Information
Q How satisfied or dissatisfied were you with..?

% Dissatisfied % Satisfied

How easy it was to


understand the information 8% 86%
you were given
Being given all the 15% 80%
information you needed

The accuracy of the 13% 79%


information you were given

The way you were kept 19% 64%


informed about progress

Base: All respondents (1,502) Source: MORI

15
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Similar groups of people tend to be more satisfied with information as with the
previous service aspects – older people, referring to health and education and those
who contact the service in person. Those contacting a local council or using the
telephone or a letter to contact public services are most dissatisfied.

How easy it Being given The The way you


was to all the accuracy of were kept
understand information the informed
the you needed information about
information you were progress
you were given
given
Satis- Dis- Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis- fied satis-
Base fied fied fied fied
% % % % % %
Type of service
Health services (612) 91 6 87 9 87 8 73 13
Education services (389) 92 3 87 9 87 6 75 13
Local council (84) 72 16 53 39 56 32 36 43
The Police (87) 78 11 69 26 63 21 52 38
Transport (85) 71 15 68 24 59 29 33 31
Method of contact
Personally (812) 87 6 83 11 82 10 67 15
Telephone (525) 84 10 75 20 74 18 57 25
Letter (78) 82 15 77 19 74 12 63 28
Email (39) 89 5 81 14 83 14 66 26

Source: MORI

16
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Driver 4: Professionalism


There is high satisfaction with the competency of staff and most people feel they
were treated fairly and their privacy was protected. Slightly fewer people consider
the service reliable.

Professionalism
Q How satisfied or dissatisfied were you with..?

% Dissatisfied % Satisfied

How competent staff were 12% 83%

How fairly you were 11% 83%


treated

The way your privacy was 4% 79%


protected

How reliable the service was 16% 78%

Base: All respondents (1,502) Source: MORI

17
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Younger people, those contacting the police, local council or transport and those
who use a letter or telephone to contact public services are most dissatisfied.

How How fairly The way How reliable


competent you were your privacy the service
staff were treated was was
protected
Satis- Dis- Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis- fied satis-
Base fied fied fied fied
% % % % % %
Type of service
Health services (612) 91 6 90 6 92 2 87 9
Education services (389) 89 7 86 6 76 4 84 8
Local council (84) 53 39 64 22 55 5 40 40
The Police (87) 77 19 75 22 68 10 62 28
Transport (85) 68 19 65 20 48 4 58 37
Method of contact
Personally (812) 88 8 86 8 80 5 83 12
Telephone (525) 78 17 78 15 79 3 72 23
Letter (78) 70 17 67 15 62 8 64 23
Email (39) 60 21 75 18 79 0 70 21

Source: MORI

18
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Driver 5: Staff attitude


Public services staff generally come across as polite and friendly and people are
satisfied with how much time they spend with them. Relatively speaking, people are
slightly less happy with how sympathetic staff are to their needs, although this still
amounts to nearly eight out of ten people who are happy.

Staff attitude
Q How satisfied or dissatisfied were you with..?

% Disagree % Agree

How polite staff were 6% 89%

How friendly staff were 6% 88%

The amount of time staff


spent with you 10% 81%

How sympathetic staff


were to your needs 14% 78%

Base: All respondents (1,502) Source: MORI

19
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Again, older people and those making personal contact tend to be more satisfied.
Additionally to health and education, police staff are also praised for their attitude.

How polite How Amount of How


staff were friendly staff time staff sympathetic
were spent with staff were to
you your needs
Satis- Dis- Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis- fied satis-
Base fied fied fied fied
% % % % % %
Type of service
Health services (612) 94 4 94 4 88 8 85 8
Education services (389) 92 4 91 4 85 7 82 8
Local council (84) 73 14 69 18 63 19 48 39
The Police (87) 89 8 85 8 79 14 76 16
Transport (85) 72 12 70 14 58 10 57 27
Method of contact
Personally (812) 92 5 92 5 84 9 83 10
Telephone (525) 87 8 85 9 79 12 73 17
Letter (78) 70 6 70 6 68 10 63 22
Email (39) 70 14 72 13 74 11 61 14

Source: MORI

20
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Key Driver 6: Physical environment


The general public are largely satisfied with the way public services premises and
staff look. In particular, very few are dissatisfied with staff appearance.

Physical environment
Q How satisfied or dissatisfied were you with..?

% Disagree % Agree

The appearance of staff 2% 92%

How clean the premises 8% 83%


were

How comfortable the


premises were 8% 80%

Base: All who contacted a public service in person (957) Source: MORI

21
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Younger people and those from higher social class are more dissatisfied with the
physical environment overall. Additionally, those using transport are particularly
dissatisfied with the cleanliness and the comfort of premises (though note the small
base size here).

The appearance How clean the How


of staff premises were comfortable the
premises were
Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis-
Base fied fied fied
% % % % % %
Type of service
Health services (441) 94 1 88 9 87 7
Education services (236) 92 5 92 3 88 6
Local council (30) 84 3 77 0 70 5
The Police (38) 83 4 39 2 44 4
Transport (49) 82 2 61 21 56 25
Method of contact
Personally (812) 93 2 85 8 82 8
Telephone (114) 88 1 71 4 69 8
Letter (15) 67 6 70 13 56 7
Email (9) 64 0 45 5 80 0

Source: MORI

22
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Factor 7: Access
Generally, people find it easy to access public services. However, there is slightly
less satisfaction with the choice of different ways to access the service and with the
opening hours.

Access
Q How satisfied or dissatisfied were you with..?

% Disagree % Agree

How easy it was to 10% 85%


access the service

The opening hours of


the service 10% 76%

The choice of different


ways you had to access 9% 71%
the service

Base: All respondents (1,502) Source: MORI

23
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Younger people and those from higher social class are most dissatisfied with access
issues (perhaps because they have less free time, or because they have higher
expectations), as are those using telephone, letter or email to contact public services
instead of a personal contact.

How easy it was The opening The choice of


to access the hours of the different ways
service service you had to
access the
service
Satis- Dis- Satis- Dis- Satis- Dis-
fied satis- fied satis- fied satis-
Base fied fied fied
% % % % % %
Type of service
Health services (612) 88 7 80 10 70 7
Education services (389) 89 5 80 6 80 4
Local council (84) 74 16 71 16 70 14
The Police (87) 76 18 56 14 65 19
Transport (85) 78 14 66 14 66 14
Method of contact
Personally (812) 88 6 81 9 69 6
Telephone (525) 81 15 73 13 73 13
Letter (78) 77 16 63 11 75 6
Email (39) 72 16 68 15 90 3

Source: MORI

24
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Summary of Performance on Key Drivers


We created an overall average of performance for each factor and plotted it against
the impact it has on satisfaction with public services, as shown on the chart below.
The higher up on the chart a key driver is, the more satisfied people are with it and
the further along a key driver is, the more impact it has on satisfaction with public
services.

People are most satisfied with staff attitude, which has a relatively smaller impact on
their overall satisfaction with public services. Delivery has a comparatively larger
impact on overall satisfaction with public service but people are less satisfied with
this aspect of public services – better information and timeliness are also priorities
for improvement.

Performance on Key Drivers of Satisfaction


Average weighted satisfaction score
(5= very satisfied, 1 = very dissatisfied)
4.5

Staff attitude
4.3
Professionalism

4.1
Delivery

Information
3.9 Timeliness

3.7

3.5
5 10 15 20 25 30 35
% impact of driver
Base: All respondents (1,502) Source: MORI

25
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Additional Key Drivers Analysis


Key driver analysis was also carried out on specific subgroups: those who have
contacted education services, those who have contacted health services and those
who have contacted other services (primarily local government, the police and
transport).

In each case, the same factors, in broadly the same order, are key drivers of
satisfaction. This suggests that even though there will be specific issues particular to
each service, these five factors are important across the board.

Education services

Main elements Drivers


The final outcome
The way the service kept its promises Delivery
The way the service handled any problems
34
%
Initial wait
How long it takes overall Timeliness
Number of times had to contact the service 23
%
Accuracy
Comprehensiveness Information 17% Satisfaction
Being kept informed about progress
with service
%
Competent staff 14
Being treated fairly Professionalism
%
12

Polite and friendly staff


How sympathetic staff were to your needs Staff attitude

This model explains 62% of the


variation in satisfaction Source: MORI

26
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Health services

Main elements Drivers


The final outcome
The way the service kept its promises Delivery
The way the service handled any problems
28
%
Initial wait
How long it takes overall Timeliness
Number of times had to contact the service 24
%
Accuracy
Comprehensiveness Information 19% Satisfaction
Being kept informed about progress
with service
%
Competent staff 15
Being treated fairly Professionalism

%
13
Polite and friendly staff
How sympathetic staff were to your needs Staff attitude

This model explains 59% of the


variation in satisfaction Source: MORI

Other services

Main elements Drivers

The final outcome


The way the service kept its promises Delivery
The way the service handled any problems
31
%
Initial wait
How long it takes overall Timeliness
Number of times had to contact the service 23
%

Competent staff
Professionalism 18% Satisfaction
Being treated fairly
with service
%
Accuracy 18
Comprehensiveness Information
Being kept informed about progress
%
10

Polite and friendly staff


How sympathetic staff were to your needs Staff attitude

This model explains 72% of the


variation in satisfaction Source: MORI

27
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Expectations of the Public and


Private Sector
What are the perceptions of public services?
Where possible, the perceptions and expectations of public services among the
general public should be placed in context. As part of this research, two
comparisons can be offered – one is between perceptions and expectations of
public services compared with private sector companies and the other is looking at
changes in expectations of public services over time.

People think the private sector is better at providing services than the public sector
by a margin of more than 3:2. Just as many though, think there is no difference
between the two. Similarly, more people say the private sector has improved over
the last five years than say the same of the public sector. The middle classes and
younger people are most likely to praise the private sector over the public sector
(although older people are more likely to think both public and private services have
got worse over the last five years).

Public and private sector performance


Q I would like you to think about public sector organisations like local councils, schools, and
hospitals, and private sector companies like shops and banks. Which do you think is better at
providing services – public sector organisations or private sector companies – or do you think
there is no difference?

Don’t know
6% Public sector better
20%

No difference 38%

36% Private sector better

Base: All respondents (1,502) Source: MORI

28
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Performance of Public Services

Q On balance, do you think public services like local councils, schools or hospitals /
private sector companies like shops or banks have got better or worse over the
last five years, or do you think they have stayed the same?

% Better % Stayed the same % Worse % Don’t know

Public services (1998) 24 38 33 5

Public services 29 29 39 3

Private sector 32 38 27 3

Base: 2004 - all respondents (1,502). 1998 – all respondents (5,064) Source: MORI

Since 1998, the proportion who say public services exceed their expectations has
risen by five points. Unfortunately, there has been an even bigger, significant rise in
those who say public services fail to meet their expectations (up 11 points). People
are more likely to say the private sector meets their expectations (though this in
itself does not prove whether people have higher or lower expectations of the
private sector over the public sector).

Expectations of Public Services


Q Thinking generally about what you expect of public services like local councils,
schools or hospitals/private sector companies like shops or banks, would you say
they greatly exceed or slightly exceed your expectations, are about what your
expect, fall slightly short or fall a long way short of your expectations?

% Exceed % About what you expect % Fall short % Don’t know

Public services (1998) 5 51 40 4

Public services 10 38 51 1

Private sector 11 59 28 3

Base: 2004 - all respondents (1,502). 1998 – all respondents (5,064) Source: MORI

29
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

What are the expectations of responsiveness?


There is an ongoing debate over whether how people perceive public services may
be caused by changes in their expectations of public services. In particular, some
believe that expectations have risen over the past few years, which may lead to
greater dissatisfaction with the services currently provided, regardless of changes in
service quality.

One of the ways we are able to measure any changes in expectations is to ask what
people perceive to be an acceptable wait when they contact public and private
sector services. Some shifts towards lower expectations of public services can be
observed, as shown on the chart below.

When contacting a service in person, but without an appointment, or when a letter


is sent, people have lower expectations of the speed of response public services
should provide now then they did in 1999. They also expect a faster response from
private sector companies than from public services.

Q When you visit your local council or a government department / a private


sector company like a shop or a bank without an appointment, how long do
you think it is reasonable to wait before someone deals with your query or
problem?

Public sector Public sector Private sector


1999
Base (1,003) Base (753) Base (749 )
% % %
Immediately 4 5 11
0-5 minutes 9 9 18
Over 5 minutes – 10 minutes 19 21 25
Over 10 minutes – 15 24 20 14
minutes
Over 15 minutes – 20 13 10 8
minutes
Over 20 minutes – 25 3 2 1
minutes
Over 25 minutes – 30 16 11 8
minutes
Over 30 minutes 8 14 11

Q. When you send a letter to your local council or government department / a


private sector company like a shop or a bank, how long do you think it is
reasonable to wait for your letter to be answered?
% % %
1-3 working days 9 9 21
Over 3 working days – 1 week 62 55 59
Over 1 week – 2 weeks 23 27 16
Over 2 weeks – 3 weeks 4 4 2
Over 3 weeks – a month 1 1 *
Over a month * 2 *

Source: MORI

30
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

When the contact is made by email, the private sector is again expected to provide a
speedier response, but the expectations of how quickly public services should
respond have remained the same between 1999 and 2004.

Q. When you send an e-mail to your local council or government department / a


private sector company like a shop or a bank, how long do you think it is
reasonable to wait for your email to be answered?
Public sector Public sector Private
1999 sector
Base: all with email Base: all who use emails
(116) (476) (475)
% % %
Answered immediately 1 4 4
Less than one hour 6 4 5
Over one hour – 2 hours 3 5 6
Over 2 hours – 4 hours 7 5 5
Over 4 hours – 8 hours 5 3 3
Over 8 hours – 24 hours 27 25 35
Over 1 – 2 working days 24 26 26
Over 2 – 3 working days 13 9 6
Over 3 – 5 working days 4 11 1
Over one week 5 2 1

Source: MORI

The expectations of the speed of response when contacting a public or private


sector organisation by telephone are similar, without any significant changes in
expectations of public services between 1999 and 2004.

Q. When you contact your local council or government department/a private


sector company like a shop or a bank, by telephone, how long do you think it
is reasonable to wait for your call to be answered?

Public sector Public sector Private


1999 sector
Base (1,003) Base (753) Base (749 )
% % %
Within 15 seconds (5 rings) 20 25 23
Within 30 seconds (10 31 24 25
rings)
Within 1 minute (20 rings) 18 18 20
Over 1 – 2 minutes 13 13 12
Over 2 minutes 16 19 19
Don’t know 2 2 1

Source: MORI

31
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Methods of Contact
Personal contact and telephone are the two most frequent methods of contact
people used to get in touch with public services. Only a small proportion wrote
letters or emails. Telephone and personal contact are also the two most preferred
methods, but this time people would rather use the phone than face-to-face contact.

People contact health, education and transport services in person, but they are more
likely to use the telephone to contact the police and transport. A letter is used
primarily to contact a local council.

Telephone is the preferred method for contacting health services, the police and
local councils, while more people would rather contact education and transport
services in person. Younger people and the middle classes are slightly more likely to
prefer email contact with public services.

Methods of contact
Q Still thinking about your contact with <service>, could you tell me how you
were in contact with the service?
Q And if you had to contact <service> again, how would you prefer to do it?

Actual
Preferred

55%
In person
46%
36%
Telephone
52%
5%
Letter
5%

2%
E-mail*
3%

Base: All respondents (1502). *All who use email (392) Source: MORI

32
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

Conclusions
This study has derived a generic model of what drives satisfaction with public
services. This is useful as it provides a framework for the public sector within which
it can focus its efforts on improving customer service: when an organisation
performs well on each of the key drivers, it will substantially improve its chances of
receiving high satisfaction scores for the services it provides.

The key drivers, in order of their impact, are:

1. Delivery - the service delivers the outcome it promised and manages to deal
with any problems that may arise.

2. Timeliness – the service responds immediately to the initial customer


contact and deals with the issue at the heart of it quickly and without
passing it on between staff.

3. Professionalism – staff are competent and treat customers fairly.

4. Information - the information given out to customers is accurate and


comprehensive and they are kept informed about progress.

5. Staff attitude - staff are friendly, polite and sympathetic to customers’ needs.

By looking at the average performance for each factor and plotting it against the
impact it has on satisfaction with public services, we are also able to indicate what
the priorities for improvement should be. As described earlier, on the chart below
the higher up a driver is, the higher its average satisfaction score, and the further to
the right the more impact it has on overall satisfaction with public services.

Performance on Key Drivers of Satisfaction


Average weighted satisfaction score
(5= very satisfied, 1 = very dissatisfied)
4.5

Staff attitude
4.3
Professionalism

4.1
Delivery

Information
3.9 Timeliness

3.7

3.5
5 10 15 20 25 30 35
% impact of driver
Base: All respondents (1,502) Source: MORI

33
The Drivers of Satisfaction with Public Services for the Office of Public Services Reform

This indicates that while people are generally satisfied with the attitude and
professionalism public sector staff demonstrate in their jobs, delivery in particular,
but also timeliness (and to an extent information provision), should be priorities for
public services.

Of course, given the very nature of public services, there are inevitably going to be
occasions when the public sector does not provide what the customer wants, or in
cases of regulatory or enforcement services can actually provide a ‘negative’
outcome (from the customer’s point of view). However, even in these situations,
emphasis can be placed on the other components of “delivery” – keeping promises
(which requires expectations of the outcome to be well managed at the outset), and
dealing effectively with any problems that may arise.

Following on from this, good performance on the remaining four drivers of


satisfaction can also begin to make up for dissatisfaction with the final outcome.
Even negative outcomes can be handled without wasting the customers’ time, by
friendly and polite staff who are competent at what they do and provide accurate
and comprehensive information.

The drivers identified in this research form a useful starting point for managers and
policy makers to direct service improvement where it will have the biggest impact
on satisfaction. Additionally, individual public services should of course conduct
research to identify the specific drivers for their services, but these five factors
should be applicable to all.

34
Appendices
Key Drivers survey for Office of Public Services Reform

Note on Weighting

The results in the computer tables are weighted to reflect the demographic
profile of the general public aged 16+ in Great Britain based on the Census 2001
data. In particular, weighting was carried out to match the profile in terms of age,
country/region and work status.

A total of 1,502 interviews were conducted, 287 of which were ‘education


booster’ interviews. These interviews followed special routing in the
questionnaire which meant that respondents who declared a contact with
education services in the last 12 months were then forced to answer questions on
satisfaction with the service and importance of different aspects (Q14 – Q71)
based on that experience with an education service, regardless of other services
they may or may not have used.

This meant that the data from Q14 – Q71 had to be additionally weighted to
correct for the effect of extra interviews conducted about education. In practice,
this meant reducing the overall proportion of education interviews from an initial
25% within the sample to 10%. Only the results of Q14 – Q71 were affected by
downweighting education interviews.
Key Drivers survey for Office of Public Services Reform

Statistical Reliability

The sample tolerances that apply to the percentage results for the MORI/OPSR
survey are given in the table below. This table shows the possible variation that
might be anticipated because a sample, rather than the entire population of the
general public, was interviewed. As indicated, sampling tolerances vary with the
size of the sample and the size of the percentage results.

For example, on a question where 50% of the people in a sample of 1,502 people
respond with a particular answer, the chances are (95 in 100) that this result
would not vary by more than three percentage points, plus or minus, from a
complete coverage of the general public using the same procedures.

Approximate sampling tolerances applicable to percentages at or near these levels


(Adjusted to take account of weighting)
10% or 90% 30% or 70% 50%
Size of sample on which survey ± ± ±
result is based
All respondents (1,502) 1.5 2.3 2.5
Source: MORI

Tolerances are also involved in the comparison of results from different parts of
the sample. A difference, in other words, must be of at least a certain size to be
considered statistically significant. The following table is a guide to the sampling
tolerances applicable to comparisons.

Differences required for significance at or near these percentages


10% or 90% 30% or 70% 50%
Size of sample on which
survey result is based
All satisfied with service (1,143) vs.
3.8 5.8 6.3
all dissatisfied (311)
All female respondents (771) vs. all
3.1 4.7 5.2
male respondents (731)
Health respondents (612) vs.
3.8 5.8 6.4
Education respondents (389)
All Midlands respondents (381) vs. 5.9 9.0 9.9
all respondents from Scotland (134)
Source: MORI
Key Drivers survey for Office of Public Services Reform

Social Class Definitions


The grades detailed below are the social class definitions as used by the Institute
of Practitioners in Advertising, and are standard on all surveys carried out by
MORI (Market & Opinion Research International Limited).

Social Grades
Social Class Occupation of Chief Percentage of
Income Earner Population

Higher managerial,
A Upper Middle Class administrative or
3.0
professional

Intermediate managerial,
B Middle Class administrative or
20.8
professional

Supervisor or clerical and


junior managerial,
C1 Lower Middle Class
administrative or
27.3
professional

C2 Skilled Working Class Skilled manual workers 21.2

Semi and unskilled


D Working Class
manual workers 17.4

Those at the lowest State pensioners, etc, with


E
levels of subsistence no other earnings 10.3

You might also like